Characterisation of Luvisol compaction under two different tillage systems and field traffic zones by assessing soil mechanical properties

Compaction of arable soils is a consequence of tillage systems and agricultural machinery traffic year after year. Its negative effects on crop production and on the environment have been put into evidence by several studies. However, soil compaction is a complex phenomenon and the understanding of the involved mechanisms related to agricultural practices still remains limited. This contribution aims to study the influence of the interaction between traffic intensity and tillage system on soil compaction. Soil samples were taken from topsoil (0.07-0.25 m), plough pan (0.30-0.35 m) and subsoil (0.35 – 0.52 m), on plots under long-term reduced tillage (RT) and conventional tillage (CT). For each tillage system, intensive traffic zones (IT) and non-intensive traffic zones (NT) were considered. Swelling index (Cs), compression index (Cc), precompression stress (Pc) obtained by oedometer test, porosity (n) and water content obtained by gravimetric determination were chosen to characterise the soil mechanical properties. An analysis of covariance (ANCOVA) was performed to study the effect of the depth, the tillage and the traffic intensity on the variables measured, with the water content as covariable. The results show that, after ten years of reconversion from CT to RT, the plough pan is still present in RT and its compaction appears as important as in CT ( nRT-30cm = 36.9% , nCT-30cm = 38.0%, p-value = 0.098). In subsoil, the compression index was high in CT, as well as in RT (CcRT = 0.150 kPa-1, CcCT = 0.148 kPa-1, p-value = 0.617), involving that this layer remains susceptible to compaction under heavy loads. Moreover, the mean value of the precompression stress (meanPc = 92±34 kPa) remains lower than stresses induced by heavy machines such as beet harvesters. The results also show that the presence of two traffic zones induces a spatial heterogeneity in the field (CcIT = 0.138 kPa-1; CcNT = 0.154 kPa-1, p-value = 0.031). These main results could be used in computational modelling to develop decision support systems to mitigate the effects of soil compaction.